This project aims to identify dietary, hormonal, and genetic factors related to prostate cancer in the Health Professionals Follow-Up Study (HPFS). In spite of the slow progression rates of most prostate cancers, a subset of patients will experience a more aggressive and generally fatal course. The heterogeneity in biologic potential for progression among tumors is likely related to acquired molecular characteristics. Thus, we will try to link dietary and hormonal factors that are related to aggressive behavior in prostate cancer to specific molecular characteristics that determine biologic potential, including cell proliferation, differentiation, apoptosis, inflammation, and angiogenesis. We will also examine whether identified dietary and other modifiable risk factors can, in the post-diagnostic period, influence the risk for PSA relapse among men treated with apparently organ-confined prostate cancer. Because our ultimate goal is to provide feasible approaches for prevention, in addition to many aims geared towards understanding the disease (hormones, genetic factors, total energy intake), the focus of many of our aims is on modifiable factors (aspirin, calcium, omega-3 fatty acids, lycopene (tomatoes), and vitamin E). Among 47,000 HPFS men free of cancer at baseline in 1986, we anticipate 4,124 new cases of prostate cancer by 2004, including 573 metastatic cases. We also plan to acquire tumor blocks from 1,732 prostatectomy cases. The sources of the exposure data in the HPFS are (1) questionnaire, including diet, (2) plasma samples (insulin-like growth factors, sex hormones, vitamin D, carotenoids), and (3) blood and buccal cell DNA for MnSOD, vitamin D and androgen receptor CAG polymorphisms. The outcome data will be based on (1) medical record and pathology report review for initial diagnosis and relapse (PSA failure), and (2) tissue blocks for microvessel density, VEGF, COX-2, PTEN loss, proliferation, and apoptosis. Stratified analysis and multivariate analysis will be used to control for potential confounding factors. This project is likely to yield important new findings that may help our understanding of modifiable risk factors for prostate cancer incidence, progression, and relapse.